Event Segmentation of Agent Interactions: Comparing the Whole with Its Parts

How do observers perceptually organize the events of individual agents when observing interactions among them? Do they readily perceive all events? Do they selectively perceive some events but not others? Do they see events overlooked by observers focusing on the individual agents? To explore these questions, participants viewed the Heider and Simmel (1944) animation, which shows three moving figures and elicits strong impressions of interacting agents. Participants in the default condition segmented the animation into meaningful events. Those in focus conditions did likewise, but focusing on one of the figures. Results indicate that participants in the default condition disregarded many events identified in the focus conditions, but identified only one event missed by focus-condition participants. These findings suggest that observers of interactions do not encode all events or gain additional insight by “seeing the big picture”; rather, they selectively perceive some events at the cost of overlooking others